Gemcitabine is an antimetabolite with broad solid tumor activity in patients. In addition, it is a potent radiation sensitizer. During the previous grant period, we investigated the metabolic and repair pathways important for cytotoxicity and radiosensitization with gemcitabine. Mechanistic studies demonstrated that inhibition of ribonucleotide reductase, mediated by the diphosphate of gemcitabine, was responsible primarily for inhibition of DNA synthesis, whereas incorporation of the analog into DNA contributed more to cytotoxicity. In addition, ribonucleotide reductase-mediated decrease in deoxynucleotides correlated strongly with radiosensitization. We tested the hypothesis that the imbalance in deoxynucleotides led to misincorporation of nucleotides into DNA which, if not repaired prior to irradiation, resulted in radiosensitization. Results demonstrated that gemcitabine produced mismatches in DNA, which occurred only at radiosensitizing concentrations and persisted only after irradiation. Radiosensitization with gemcitabine was enhanced by mismatch repair deficiency, which increased mismatches in DNA, and suppression of the p53-inducible ribonucleotide reductase subunit p53R2, which prolonged the deoxynucleotide pool imbalances. In contrast, mismatch repair deficiency decreased cytotoxicity with gemcitabine. We now propose that these mechanisms can be generalized to other antimetabolites that produce alterations in deoxynucleotide pools and function as radiosensitizers, such as hydroxyurea (ribonucleotide reductase inhibitor), and fluorodeoxyuridine, methotrexate and pemetrexed (thymidylate synthase inhibitors) to produce the first unifying hypothesis for antimetabolite radiosensitization. We will also evaluate cytotoxicity and radiosensitization of shRNA-mediated suppression of the enzymes targeted by these antimetabolites, and compare the in vivo antitumor efficacy of shRNA enzyme suppression with antimetabolites alone or with ionizing radiation. New data indicate that late- occurring DNA double strand breaks which are unable to be repaired by homologous recombination are responsible for radiosensitization with gemcitabine. The DNA damage and repair pathways, which may include ATM, ATR and homologous recombination, that produce radiosensitization will be evaluated. Preliminary results indicate that suppression of either thymidylate synthase or the R2 subunit of ribonucleotide reductase with shRNA is at least as effective as antimetabolites in producing radiosensitization. Mechanistic studies will aid in developing a general hypothesis for cytotoxicity and radiosensitization with the antimetabolites and shRNAs. In vivo animal studies will explore dosing antimetabolites or shRNA for radiosensitization based on the amount of drug or shRNA that will decrease dNTPs and produce misincorporated deoxynucleotides in tumor cell DNA. We further hypothesize that suppression of R2 in vivo will produce superior radiosensitization with lower normal tissue toxicity. These mechanism-based studies have high potential for translation to clinic to improve radiosensitization protocols for patients.